foxai.explainer.computer_vision package

Subpackages

• foxai.explainer.computer_vision.algorithm package
  • Submodules
  • foxai.explainer.computer_vision.algorithm.conductance module
    • LayerConductanceCVExplainer
      • LayerConductanceCVExplainer.calculate_features()
      • LayerConductanceCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.deconv module
    • BaseDeconvolutionCVExplainer
      • BaseDeconvolutionCVExplainer.calculate_features()
      • BaseDeconvolutionCVExplainer.create_explainer()
    • DeconvolutionCVExplainer
      • DeconvolutionCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.deeplift module
    • BaseDeepLIFTCVExplainer
      • BaseDeepLIFTCVExplainer.calculate_features()
      • BaseDeepLIFTCVExplainer.create_explainer()
    • DeepLIFTCVExplainer
      • DeepLIFTCVExplainer.create_explainer()
    • LayerDeepLIFTCVExplainer
      • LayerDeepLIFTCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.deeplift_shap module
    • BaseDeepLIFTSHAPCVExplainer
      • BaseDeepLIFTSHAPCVExplainer.calculate_features()
      • BaseDeepLIFTSHAPCVExplainer.create_explainer()
    • DeepLIFTSHAPCVExplainer
      • DeepLIFTSHAPCVExplainer.create_explainer()
    • LayerDeepLIFTSHAPCVExplainer
      • LayerDeepLIFTSHAPCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.gradcam module
    • BaseGradCAMCVExplainer
      • BaseGradCAMCVExplainer.calculate_features()
      • BaseGradCAMCVExplainer.create_explainer()
    • GuidedGradCAMCVExplainer
      • GuidedGradCAMCVExplainer.calculate_features()
      • GuidedGradCAMCVExplainer.create_explainer()
    • LayerBaseGradCAM
      • LayerBaseGradCAM.activations
      • LayerBaseGradCAM.forward()
      • LayerBaseGradCAM.get_saliency_map()
      • LayerBaseGradCAM.gradients
    • LayerGradCAMCVExplainer
      • LayerGradCAMCVExplainer.calculate_features()
      • LayerGradCAMCVExplainer.create_explainer()
    • LayerGradCAMObjectDetection
      • LayerGradCAMObjectDetection.forward()
    • LayerGradCAMObjectDetectionExplainer
      • LayerGradCAMObjectDetectionExplainer.forward()
  • foxai.explainer.computer_vision.algorithm.gradient_shap module
    • BaseGradientSHAPCVExplainer
      • BaseGradientSHAPCVExplainer.calculate_features()
      • BaseGradientSHAPCVExplainer.create_explainer()
    • GradientSHAPCVExplainer
      • GradientSHAPCVExplainer.create_explainer()
    • LayerGradientSHAPCVExplainer
      • LayerGradientSHAPCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.guided_backprop module
    • BaseGuidedBackpropCVExplainer
      • BaseGuidedBackpropCVExplainer.calculate_features()
      • BaseGuidedBackpropCVExplainer.create_explainer()
    • GuidedBackpropCVExplainer
      • GuidedBackpropCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.input_x_gradient module
    • BaseInputXGradientSHAPCVExplainer
      • BaseInputXGradientSHAPCVExplainer.calculate_features()
      • BaseInputXGradientSHAPCVExplainer.create_explainer()
    • InputXGradientCVExplainer
      • InputXGradientCVExplainer.create_explainer()
    • LayerInputXGradientCVExplainer
      • LayerInputXGradientCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.integrated_gradients module
    • BaseIntegratedGradientsCVExplainer
      • BaseIntegratedGradientsCVExplainer.calculate_features()
      • BaseIntegratedGradientsCVExplainer.create_explainer()
    • IntegratedGradientsCVExplainer
      • IntegratedGradientsCVExplainer.create_explainer()
    • LayerIntegratedGradientsCVExplainer
      • LayerIntegratedGradientsCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.lrp module
    • BaseLRPCVExplainer
      • BaseLRPCVExplainer.add_rules()
      • BaseLRPCVExplainer.calculate_features()
      • BaseLRPCVExplainer.create_explainer()
    • LRPCVExplainer
      • LRPCVExplainer.create_explainer()
    • LayerLRPCVExplainer
      • LayerLRPCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.noise_tunnel module
    • BaseNoiseTunnelCVExplainer
      • BaseNoiseTunnelCVExplainer.calculate_features()
      • BaseNoiseTunnelCVExplainer.create_explainer()
    • LayerNoiseTunnelCVExplainer
      • LayerNoiseTunnelCVExplainer.create_explainer()
    • NoiseTunnelCVExplainer
      • NoiseTunnelCVExplainer.create_explainer()
  • foxai.explainer.computer_vision.algorithm.occlusion module
    • OcclusionCVExplainer
      • OcclusionCVExplainer.calculate_features()
  • foxai.explainer.computer_vision.algorithm.saliency module
    • SaliencyCVExplainer
      • SaliencyCVExplainer.calculate_features()
  • Module contents
• foxai.explainer.computer_vision.object_detection package
  • Submodules
  • foxai.explainer.computer_vision.object_detection.base_object_detector module
    • BaseObjectDetector
      • BaseObjectDetector.forward()
      • BaseObjectDetector.preprocessing()
      • BaseObjectDetector.training
  • foxai.explainer.computer_vision.object_detection.types module
    • ObjectDetectionOutput
      • ObjectDetectionOutput.logits
      • ObjectDetectionOutput.predictions
      • ObjectDetectionOutput.saliency_maps
    • PredictionOutput
      • PredictionOutput.bbox
      • PredictionOutput.class_name
      • PredictionOutput.class_number
      • PredictionOutput.confidence
  • foxai.explainer.computer_vision.object_detection.utils module
    • resize_image()
  • Module contents

Submodules

foxai.explainer.computer_vision.model_utils module

File contains functions to modifiy DNN models.

foxai.explainer.computer_vision.model_utils.get_last_conv_model_layer(model: Module) → Module

Get the last convolutional layer from the torch model.

Parameters:

model – torch.nn.Module

Returns:

The last convolutional layer of the model.

Raises:

ValueError if the model does not contain convolutional layers. –

foxai.explainer.computer_vision.model_utils.modify_modules(model: Module) → Module

Modify modules of given model.

Function iterates over all modules and sets property inplace to False for every torch.nn.ReLU activation function.

Parameters:

model – Neural network object to be modified.

Returns:

Modified neural network object.

Module contents